Nesting Titanosaurs from Auca Mahuevo and Adjacent Sites

TEN Nesting Titanosaurs from Auca Mahuevo and Adjacent Sites UNDERSTANDING SAUROPOD REPRODUCTIVE BEHAVIOR AND EMBRYONIC DEVELOPMENT Luis M. Chiappe, Frankie Jackson, Rodolfo A. Coria, and Lowell Dingus T housands of sauropod egg clutches, some containing eggs with exquisitely preserved embryonic bone and integument, have been discovered in the Late Cretaceous nesting site of Auca Mahuevo (Chiappe et al. 1998, 2000, 2001, 2004; Dingus et al. 2000; Chiappe and Dingus, 2001; Coria et al. 2002) and adjacent localities in northwestern Patagonia, Argentina (fig. 10.1). Five expeditions to this extraordinary area (1997, 1999, 2000, 2001, and 2002) have yielded a wealth of information for understanding the prehatching development, the nesting structure, the egg morphology and malformation, and the reproductive behavior of these dinosaurs. Cranial characters of the in ovo embryos allowed the identification of the eggs as those of titanosaurs (Chiappe et al. 1998, 2001; Salgado et al. 2005). Textural differences in the sediments containing some clutches have illuminated aspects of the nest structure of these animals (Garrido et al. 2001; Chiappe et al. 2004). Microstructural studies have expanded our understanding of the eggshell variation (Grellet-Tinner et al. 2004) and incidence of egg malformation (Jackson et al. 2001, 2004) within a titanosaur population. Detailed mapping of clutch spatial distribution and egg-bed stratigraphic position, together with studies of their sedimentary context, have provided the basis for inferring aspects of the nesting behavior of these dinosaurs (Chiappe et al. 2000; Chiappe and Dingus 2001). Combining taxonomic constraint with extensive sampling, research at Auca Mahuevo and its adjacent localities offers the clearest picture to date of sauropod reproduction and embryonic development. In this chapter, we summarize the major developments of this research program and discuss their significance in light of previous interpretations of the reproductive biology of these colossal dinosaurs. GEOLOGICAL SETTING AND ASSOCIATED DINOSAUR FAUNA Auca Mahuevo lies approximately 120 km northwest of the city of Neuquén in the homonymous Argentine province (fig. 10.1). Two adjacent nesting sites, Barreales Norte and Barreales Escondido, are 15 and 22 km south of Auca Mahuevo, respectively. These three sites occur within an 85-m-thick sequence of sandstone, 285

FIGURE 10.1. Map of the province of Neuquén (Argentina) indicating the location of Auca Mahuevo. Barriales Norte and Barreales Escondido are 15 and 22 km south of Auca Mahuevo, respectively. siltstone, and mudstone of the Anacleto Formation (fig. 10.3), one of the lithostratigraphic units of the fossiliferous Cenomanian Campanian Neuquén Group (Ramos 1981; Legarretta and Gulisano 1989; Ardolino and Franchi 1996; Leanza 1999; Dingus et al. 2000). Recent paleomagnetic analysis of rocks from the lower portion of the Auca Mahuevo section containing egg-beds 1 3 established the presence of a Reversed magnetozone in the Anacleto Formation (Dingus et al. 2000). In conjunction with earlier biochronologic correlations, this magnetozone was tentatively correlated with C33R, in the early middle Campanian, between 83.5 and 79.5 million years ago (Dingus et al. 2000). Exposures at Auca Mahuevo include at least four distinct egg-bearing layers (figs. 10.2, 10.3; egg-beds 1 4), which occur in uniform mudstones representing overbank deposits on a fluvial plain (Chiappe et al. 2000). Two of these layers (egg-beds 2 and 3) can be subdivided into two horizons of eggs, separated by a few centimeters of sediments in the case of egg-bed 3 and about 1 m in the case of egg-bed 2. Egg-beds 3 and 4 are laterally continuous for at least several kilometers (Chiappe et al. 2000; fig. 10.2). Most egg clutches exhibit no discernible evidence of nest structure. However, thin sandstones representing abandoned channel and crevasse splay deposits occur within the Auca Mahuevo section, and in egg-bed 4 they occasionally preserve nesting trace fossils (Chiappe et al. 2004). Less than 1 m above the nesting structures in egg-bed 4 is a sandy red mudstone that contains a great number of sauropod tracks (Loope et al. 2000). The sauropod tracks are recognizable as thin (1-cm-thick), laterally discontinuous limy deposits that measure up to 80 cm in diameter and are oval to circular in shape. These platter-shaped features are interpreted to contain precipitates of calcium carbonate within the track depression, possibly from evaporation of standing water (Loope et al. 2000). Traceable over several kilometers, the track horizon provides an index layer useful for verifying that the underlying nesting traces occur on the upper surface of a single sandstone stratum. Additional layers of calcium carbonate precipitates, also interpreted as footprints, occur elsewhere in the Auca Mahuevo section. Clutches from egg-bed 3 occur in paleovertisols (Chiappe and Dingus 2001), recognizable by the abundance and widely varying orientation of slickensides striated surfaces produced by soil movement within the nesting ground. Vertisols today are associated with clay-rich parent materials and are widespread in regions that experience wet dry climatic 286 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

FIGURE 10.2. Aerial photograph showing egg-beds 3 and 4 at Auca Mahuevo. These egg-beds can be traced laterally for several kilometers. cycles under semiarid to subhumid environmental conditions. Similar depositional conditions have been inferred for the Late Jurassic Late Cretaceous of the northern third of Patagonia, where the climate regime has been reconstructed as warm, arid to semiarid, and with a distinct dry season (Andreis 2001). Several egg-beds also occur at Barreales Norte and Barreales Escondido. Although stratigraphic correlations between these localities and Auca Mahuevo are still preliminary, the available data suggest that these egg-beds can be traced across the vast distances that separate the three sites. Abundant geodes of pale blue celestite crystals that occur in a discrete horizon above Auca Mahuevo s egg-bed 3 are also found at approximately the same stratigraphic position above an egg-bed at Barreales Norte. A footprint layer similar to those of Auca Mahuevo also occurs at this locality at comparable stratigraphic positions. In addition, comparable thicknesses separate egg-beds at these three localities. Several fossils of adult sauropod and theropod dinosaurs have been found at Auca Mahuevo. Remains of titanosaurs have been collected from egg-bed 4 and from strata between this egg layer and egg-bed 3 (fig. 10.3). These remains are yet to be studied in detail. Among the theropods found at this site is the nearly complete skeleton of Aucasaurus garridoi (Coria et al. 2002), an abelisaurid collected from a laminated mudstone unit some 25 m above egg-bed 4. Isolated teeth comparable in NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 287

FIGURE 10.3. Composite stratigraphic section at Auca Mahuevo. Note the presence of four stratigraphically distinct beds of essentially identical titanosaur eggs. Egg-beds 2 and 3 can each be subdivided into two layers. morphology with those of dromeosaurids and the fragmentary remains of a large indeterminate theropod the size of a charcharodontosaurid were also collected from Auca Mahuevo (Coria and Arcucci 2005). EGGS, CLUTCHES, AND NESTS Eggs from all these localities exhibit similar size, shape, microstructure, and surface ornamentation as the eggs that contain diagnostic titanosaur remains. The eggs are spherical to subspherical and approximately 13 15 cm in diameter, with a tubercular surface ornamentation consisting of single, rounded nodes (fig. 10.4). The eggshell consists of a single structural layer of calcite approximately 1.3 mm thick in well-preserved samples pierced by a pore network of vertical and horizontal canals that intersect one another at the bases of the eggshell units (Grellet-Tinner et al. 2004). The overall microstructure of the eggshell is similar to that 288 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

FIGURE 10.4. (A) Auca Mahuevo clutch (Museo Carmen Funes, Plaza Huincul, Argentina; MCF-PVPH-258; quarry of egg-bed 3 illustrated in fig. 10.5) containing nearly 40 eggs. (B) and (C) Scanning electron micrograph and thin section of the Auca Mahuevo eggshell. Scale bars in B and C equal 10 micron and 1 mm, respectively. described as the ootaxon Megaloolithus patagonicus from the Anacleto Formation at Neuquén City (Calvo et al. 1997; fig. 10.4). Megaloolithus patagonicus has recently been proposed as a possible junior synonym of Megaloolithus jabalpurensis (Vianey-Liaud et al. 2003), a Late Cretaceous (Maastrichtian) oospecies from India. Megaloolithus patagonicus is also very similar to Late Cretaceous eggshells from Perú identified by Vianey-Liaud et al. (1997) as Megaloolithus pseudomamillare (Grellet-Tinner et al. 2004). Future studies are likely to synonymize many of the megaloolithid oospecies that have been named from various localities around the world. Vianey-Liaud and others (2003) took an initial step in this direction and greatly reduced the number of valid megaloolithid oospecies from India. Although it is difficult to ascertain the completeness of a fossil egg-clutch, Auca Mahuevo s clutches are composed of numerous eggs, most typically from about 20 to nearly 40 eggs (Chiappe et al. 2004; Jackson et al. 2004; fig. 10.4). Eggs are stacked one on top of the other without any internal spatial arrangement (up to three stacked layers of eggs have been described NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 289

FIGURE 10.5. Map in plan view of eggs exposed at a quarry in Auca Mahuevo s egg-bed 3. The total surface area excavated in this quarry is approximately 65 m 2.

FIGURE 10.6. Clutch maps of two sites contained within erosional surfaces of Auca Mahuevo s eggbed 3. Seventy-four and 31 randomly distributed egg clutches were mapped within 1,701 m 2 (A) and 486 m 2 (B). Even if recognition of egg-clutch boundaries is sometimes difficult (see text), these numbers provide a minimal estimate of the clutches laid on these surfaces. for some clutches). More than 500 whole eggs were quarried over a 65-m 2 surface of egg-bed 3 during the 1999, 2002, and 2002 field seasons (fig. 10.5). Spatial analysis of the eggs exposed during 1999, in conjunction with a three-dimensional map constructed from field data, revealed an unexpectedly high egg density (11 eggs/m 2 ) in this quarry. At this quarry, the boundaries of individual egg clutches are sometimes difficult to determine. In some cases, eggs occur as large accumulations that clearly represent more than one clutch (fig. 10.5). To a certain degree, such a pattern can be explained as the consequence of postburial egg displacement due to edaphic processes a substantial displacement along the friction planes of slickensides has been observed on several instances. It is also possible that the eggs were somewhat displaced by flotation prior to their final burial. Detailed stratigraphy and mapping of egg depths within this quarry also revealed two distinct levels of eggs, separated by several centimeters of sediment, which were interpreted as different egglaying events (Chiappe et al. 2000). Mapping of egg clutches exposed on erosional surfaces of egg-bed 3 produced a concentration of 74 and 31 randomly distributed egg clutches within 1,701 and 486 m 2, respectively NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 291

FIGURE 10.7. Egg-clutch map of a site exposed on an erosional surface of Auca Mahuevo s egg-bed 4. This map shows a density comparable to that recorded for egg-bed 3 (fig. 10.6). (Chiappe et al. 2000; fig. 10.6). Egg clutches on erosional surfaces are often recognized as accumulations of partially weathered eggs whose periphery (eggshell) is still vertically oriented within the substrate. Heavily weathered clutches are recognized as large accumulations of broken eggshells that fan out from a core and that are separated from other egg clutches by areas with minimal eggshell. If any, maps constructed with these criteria for recognition underestimate the number of egg clutches that were laid within these paleosurfaces because of the abovementioned causes of egg accumulation and displacement. The maximum stratigraphic thickness of strata containing egg clutches within these areas was less than 70 cm. A similar map of an erosional surface of egg-bed 4 showed a comparable high density of randomly distributed egg clutches (fig. 10.7). This high concentration of clutches provides opportunities for studying aspects of sauropod reproductive biology that are otherwise difficult to assess. For example, the discovery of abnormal, multilayered eggs in 6 of nearly 400 in situ clutches surveyed at Auca Mahuevo s egg-beds 2 and 3 (Jackson et al. 2001, 2004) provided the first assessment of the incidence of egg malformation in a sauropod population. This study detected three different types of abnormal eggshells and the fact that pathologic eggs were laid in clutches containing a majority of normal eggs. Several egg clutches from egg-bed 4 preserve evidence of nest architecture. The eggs of these clutches are similar in size, shape, and microstructure to other Auca Mahuevo eggs containing embryonic remains of titanosaur sauropods (Chiappe et al. 2001). The clutches are contained in large, subcircular to subellipti- 292 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

FIGURE 10.8. Four titanosaur nests from Auca Mahuevo s egg-bed 4. These nests consist of surface depressions whose periphery is defined by an elevated rim of structureless sand. Textural differences between the depression fill and the substrate indicate that titanosaurs laid their eggs on the surface. cal to kidney-shaped depressions in sandstone, although the depression and interstitial spaces between the eggs are filled with mudstone (fig. 10.8). All of the depressions truncate primary stratification of the host substrate and are encircled by a rim of structureless sandstone. The maximum axis of the depressions ranges in length between 100 and 140 cm and their depth is about 10 to 18 cm. These egg-filled depressions are interpreted as excavated nests on the basis of lithologic criteria (Garrido et al. 2001; Chiappe et al. 2004): (1) the depressions truncate the primary stratification of the host sandstone, (2) a massive rim surrounds the perimeter of the depression, and (3) textural differences exist between the host sandstone and the in-filling sediment. We have interpreted the structureless sand of the depression s rim as the piled debris produced during the construction of the nest and the mudstone surrounding the eggs as the result of a flooding event that is similar to most overbank deposition of the Anacleto Formation in the Auca Mahuevo section. The fact that the eggs in the recognized nests are not entombed by sandstone but by mudstone resulting from the flooding event indicates that the nesting sauropod did not bury the eggs after they were laid (Chiappe et al. 2004). All other clutches from Auca Mahuevo are likely to have been laid in similarly constructed surface nests, which are not recognizable due to the lack of textural differences between the host mudstone and the in-filling sediment. These nests could have NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 293

FIGURE 10.9. Titanosaur embryonic remains from Barreales Norte, 15 km south of Auca Mahuevo. Note the remains of limb bones covered by the eggshell of tuberculate ornamentation. been lined and covered with vegetation (Grellet- Tinner et al. 2004), a suggestion that several researchers have made for other megaloolithidtype eggs from the Late Cretaceous of France (Erben 1970; Kérourio 1981; Cousin 1997). EMBRYONIC MORPHOLOGY AND SYSTEMATICS In ovo embryonic remains have been found in two situations: (1) encased in highly cemented fragments of eggs that occur as float on erosional surfaces and (2) compressed against the bottom inner shell of in situ eggs. Despite the abundance of embryos, no remains of hatchlings or early juveniles have been discovered. So far, most embryonic remains have been collected from Auca Mahuevo s egg-bed 3 (Chiappe et al. 1998, 2001), although some have also been found at Barreales Norte (fig. 10.9) and Barreales Escondido. Patches of integument, preserved as calcitic impressions (negative and positive), are common in highly cemented egg fragments that 294 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

FIGURE 10.10. Photos and interpretive drawings of embryonic titanosaur skulls from Auca Mahuevo s eggbed 3 in left lateral view. (A) MCF-PVPH-272; (B) MCF-PVPH-263. Arrows point to the approximate location of external nares. Abbreviations: af, antorbital fenestra; an, angular; d, dentary; f, frontal; itf, infratemporal fenestra; j, jugal; la, lacrimal; m, maxilla; mf, mandibular fenestra; orb, orbit; p, parietal; pmx, premaxilla; po, postorbital; prf, prefrontal; pt, pterygoid; q, quadrate; qj, quadratojugal; scp, scleral plates; sq, squamosal; stf, supratemporal fenestra. occur on erosional surfaces. These skin impressions display a range of nonoverlapping tubercular patterns including rosettes, flowerlike arrangements, and rows of larger tubercles (Chiappe et al. 1998). Osteological remains are most typically found flattened against the bottom inner shell of in situ eggs. Cranial material is better ossified than limb material, which typically lacks ends (fig. 10.9). Some embryos are clearly larger (by as much as 25%) than others. Although it is difficult to estimate the degree of development of the embryos in comparison to the embryonic stages of extant reptiles, the basic sauropod morphogenetic plan is readily visible in the available skulls. A general Haeckelian pattern is also evident in these skulls, which in some respects resemble conditions found outside Sauropoda (e.g., jugal forming part of the ventral margin of the skull, minimally retracted nares). The anatomical information available in the handful of embryos initially prepared supported the identification of these embryos as neosauropod dinosaurs (Chiappe et al. 1998), the clade originating from the common ancestor of the Late Jurassic Diplodocus longus and the Late Cretaceous titanosaur Saltasaurus loricatus (Wilson and Sereno 1998). Originally, although synapomorphies of Sauropoda were identified among the disarticulated skulls (e.g., jugal process of the postorbital much longer than the rostrocaudal extension of the dorsal end of this bone [Chiappe et al. 1998]), the dental morphology played a key role in the taxonomic identification of the embryos: the smooth enamel (devoid of denticles) of the crowns and the pencil shape (straight margins and tapering NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 295

FIGURE 10.11. Simplified cladogram of sauropod relationships. Derived dental characters present in the Auca Mahuevo embryos support their inclusion in a Subgroup of Titanosauria that excludes the African Malawisaurus dixeyi. (After Salgado et al. 1997.) Icons from Wilson and Sereno (1998). crowns) of the teeth, derived characters of neosauropods and diplodocoids titanosaurs, respectively (Wilson and Sereno 1998; Wilson 2002), supported the placement of the embryos within Neosauropoda. However, because pencilshaped teeth are commonly interpreted as independently evolved in both diplodocoids and titanosaurs (e.g., Salgado and Calvo 1997; Wilson and Sereno 1998; Curry Rogers and Forster 2001; Wilson 2002), the presence of this condition alone was insufficient for placing the Auca Mahuevo embryos within either of these two neosauropod clades. Subsequent discoveries of embryos yielding more complete and articulated skulls (Chiappe et al. 2001; Salgado et al. 2005) (fig. 10.10) revealed additional synapomorphies of neosauropods and of the more inclusive taxon Eusauropoda (fig. 10.11) taxa more closely related to Saltasaurus loricatus than to Vulcanodon karibaensis (Wilson and Sereno 1998). For example, the new embryos display the eusauropod condition of a stepped snout, an absence of a fossa surrounding the antorbital fenestra, a rostral expansion of the quadratojugal, and a lack of contact between this bone and the squamosal (Chiappe et al. 2001; fig. 10.10). Likewise, the new embryos exhibit the neosauropod condition of a postorbital bar that is broader transversely than rostrocaudally (Chiappe et al. 2001). Most importantly, comparisons between these more recently discovered embryonic skulls and titanosaur cranial remains revealed several apparent synapomorphies of this sauropod clade. For example, the embryos exhibit the same ventral notch of the dentigerous margin, between the maxilla, jugal, and quadratojugal, as the titanosaurs Rapetosaurus krausei (Curry Rogers and Forster 2001, 2004) and Nemegtosaurus mongoliensis (Nowinski 1971), from the Late Cretaceous of Madagascar and Mongolia, respectively. They also share with this taxon and with other titanosaurs from the Late Cretaceous of Argentina (i.e., Antarctosaurus wichmannianus and an undescribed titanosaur from northwestern Patagonia) the low rostral portion of the dentary (Chiappe et al. 2001). The remarkable width of the skull roof, as inferred from the size of the frontals and parietals, and the presence of a large mandibular fenestra are other derived features shared by the embryos and the above-mentioned undescribed titanosaur from northwestern Patagonia (Coria and Salgado 1999). These embryonic skulls represent the most complete titanosaur crania. Furthermore, the embryonic material 296 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

suggests that previous skull reconstructions of these dinosaurs as camarasauroid (Salgado and Calvo 1997) are incorrect. Neither a vertical orientation of the quadrate nor the presence of broad nares separated by an elevated premaxillary nasal arch, conditions typical of camarasauroid skulls, is present in the well-preserved skulls of the Auca Mahuevo embryos (Chiappe et al. 2001). Although the recently discovered embryos have provided support for identification of the Auca Mahuevo eggs as those of titanosaurs, systematic placement of the embryos beyond Titanosauria (i.e., all sauropods more closely related to Saltasaurus loricatus than to either Brachiosaurus brancai or Euhelopus zdanskyi [Wilson and Sereno 1998]) remains problematic. This is primarily because of the paucity of cranial anatomical information available for adult titanosaurs (Salgado and Calvo 1997; Curry Rogers and Forster 2001). The presence of pencil-like teeth, so far known only for Nemegtosaurus mongoliensis (Nowinski 1971), Alamosaurus sanjuanensis (Kues et al. 1980), and saltasaurines (most recent common ancestor of Neuquensaurus australis and Saltasaurus loricatus plus all its descendants [Salgado and Calvo 1997]) among titanosaurs, provides support for the placement of the embryos within a subgroup of titanosaurs that excludes the most primitively toothed Malawisaurus dixeyi (Jacobs et al. 1993; fig. 10.11). Nonetheless, this interpretation becomes more complex in light of recent phylogenetic inferences indicating that this specialized dental condition could have evolved more than once within titanosaurs (Curry Rogers and Forster 2001). Despite these reservations, comparisons between the embryos and the best-preserved skulls of adult titanosaurs, those of Nemegtosaurus mongoliensis and Rapetosaurus krausei, suggest that dramatic transformations must have occurred during the ontogeny of these dinosaurs. The frontals and parietals became greatly reduced in size and they migrated to the dorsocaudal and caudal region of the orbit. The latter became ventrally constricted and adopted an inverted tear-shaped appearance. The rostrum became substantially enlarged, probably as a consequence of maxillary expansion, and the maxilla developed a connection with the quadratojugal, thus excluding the jugal from the ventral margin of the skull. In addition, the external nares expanded in size and migrated backward, to be relocated on top of the orbits. In addition, the Auca Mahuevo embryos have provided evidence that may potentially clarify the sequence of transformations that occurred during the long evolution of sauropods (Chiappe et al. 2001). An example of how these new developmental data could potentially elucidate aspects of sauropod evolution is provided by examining two salient features of their cranial architecture: the narial retraction and the forward rotation of the braincase. Salgado and Calvo (1997) suggested that the partial to extreme retraction of the eusauropod nares could have been evolutionarily coupled to the forward rotation of the braincase, best exemplified in diplodocids (McIntosh 1997) and some titanosaurs (Salgado and Calvo 1997; Curry Rogers and Forster 2001, 2004). Because the best-preserved embryos are exposed in lateral view, crushed against the inner shell (fig. 10.10), the exact location of the external nares cannot be directly observed. However, the location of the external nares can be inferred from the orientation of the lacrimals, which in eusauropods mark the approximate caudal end of the nares. The rostrodorsal orientation of the lacrimal suggests that the nares of the embryos opened in front of the orbit, dorsorostral to the antorbital fenestra. This position is also supported by the rostral extension of the frontals, which in the Auca Mahuevo embryos nearly reach the rostral margin of the orbit (fig. 10.10). The paucity of cranial material of adult titanosaurs has prevented determination of the location of the nares in this group of sauropods however, Rapetosaurus krausei (Curry Rogers and Forster 2004) and Nemegtosaurus mongoliensis appear to have fully retracted nares (this condition is best observed in a yet undescribed specimen of Nemegtosaurus mongoliensis housed at NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 297

the Mongolian Natural History Museum in Ulaanbataar). Although it is likely that the minimally retracted nares of the embryos migrated backward during postnatal allometric development, the rostroventral orientation of the quadrate and squamosal suggests that at that particular stage of development, the embryonic braincase was partially rotated (see Salgado and Calvo [1997] for correlations between braincase rotation and quadrate orientation). This evidence is contrary to Salgado and Calvo s (1997) hypothesis of a concerted evolution of the narial retraction and braincase rostral rotation of eusauropods (Chiappe et al. 2001). These conditions are likely to have evolved independently, although confirmation of this awaits the discovery of adult skulls with conditions resembling those of the Auca Mahuevo embryos unretracted nares and already rotated braincase. BEHAVIORAL INFERENCES Although the behavior of extinct organisms cannot be directly observed, it can be inferred when the product of an organism s activity is preserved in the fossil record (Clark et al. 1999). The in situ eggs of Auca Mahuevo and adjacent localities are the preserved physical evidence of the sauropods egg-laying behavior. The mapping and collection of eggs, clutches, and nests at these localities have led us to infer several aspects of the reproductive behavior of titanosaur sauropods (Chiappe et al. 2000; Chiappe and Dingus 2001). The high concentration of egg clutches distributed in a relatively narrow stratigraphic horizon (e.g., 70 cm in Auca Mahuevo s egg-bed 3) suggests a gregarious nesting behavior. Even if each egg-bed could preserve eggs laid during more than one closely occurring nesting season, and taking into consideration that the specifics of the gregarious behavior we have envisioned (the number of females nesting at approximately the same time and in a given season, the frequency of reproductive seasons, and other similar questions) remained unanswered, the density of eggs contained in these layers is such that the conclusion of gregariousness seems unavoidable. It is highly implausible that solitary females laid the thousands of egg clutches contained in these localities. Based on clutches that have been quarried in situ, many eggs were preserved whole, suggesting that they were buried quickly and did not sit out on the paleosurface for very long, rendering them vulnerable to natural processes of disintegration and trampling during subsequent breeding seasons. The six stratigraphically distinct egg layers (egg-beds 1 4, with egg-beds 2 and 3 each consisting of two egg levels) containing eggs of similar morphology suggest that one sauropod species nested at this site at least six separate times. A minimum of two beds of morphologically similar eggs also occurs at Barreales Norte and Barreales Escondido, supporting site fidelity for these localities as well. The most parsimonious assumption, therefore, is that all eggs were laid by the same titanosaur species. Discovery of additional embryonic remains in eggs from these egg-bearing layers will provide a means for testing this hypothesis. The discovery at Auca Mahuevo of wellpreserved nest traces provided indisputable evidence of nest construction and architecture (Chiappe et al. 2004). Sedimentological evidence showed that, contrary to most modern reptiles, titanosaurs laid eggs in excavated depressions without burying them. Although nest attendance by titanosaurs may be inferred by phylogenetic bracketing (all living archosaurs attend their nests), adult size and proximity between clutches (fig. 10.5) suggest little or no parental care of their clutches, a conclusion again supported by the lack of evidence of trampling in our quarry of egg-bed 3, where most eggs show minimal crushing. COMPARISONS TO OTHER NESTING SITES A large number of dinosaur egg localities contain clutches of eggs similar to those from Auca Mahuevo (e.g., subspherical eggs with a relatively thick eggshell comprised of a single structural 298 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

layer of calcite, with shell units well separated from each other and a tuberculate surface ornamentation) (Sahni et al. 1990; Powell 1992; Vianey-Liaud et al. 1990; Mohabey 1996; Calvo et al. 1997). Traditionally, these eggs have been classified within the Megaloolithidae category of eggshell parataxonomy and considered to have been laid by sauropod dinosaurs (Zhao 1979; Mikhailov 1991, 1997). However, Auca Mahuevo and its adjacent localities are the only sites in the world where diagnosable remains of sauropod dinosaurs have been found inside eggs, and although these embryos support the identification of some megaloolithid-type eggs (e.g., Megaloolithus patagonicus) as sauropod eggs, it would be risky to extrapolate such a conclusion to all other eggs of similar morphology. Despite this paucity of eggs containing identifiable embryonic remains at other sites, several assumptions have been made regarding the nest construction, egg-laying behavior, and physiology of sauropod dinosaurs. Inference of nest architecture typically is based on clutch geometry rather than primary lithologic attributes of the surrounding sediment (Dughi and Sirugue 1966; Kérourio 1981; Williams et al. 1984; Faccio 1990; Sahni et al. 1990; Powell 1992; Sanz et al. 1995; Mohabey 1996; Calvo et al. 1997). For example, a single layer of megaloolithid eggs from India that occurred in a 1-m 2 area was used to infer a saucer-shaped sauropod nest (Mohabey 1996, 2000). However, the margins of the nest are described as homogeneous with the host rock, with no observable lithological differences. Fossil eggs such as these provide no evidence of nest structure that results from excavation by adult dinosaurs and are, therefore, more appropriately called clutches. To the best of our knowledge, the Auca Mahuevo sauropod nests provide the only documentation of sauropod nest architecture based on lithologic attributes (Chiappe et al. 2004). A number of egg-laying behaviors have been attributed to sauropod dinosaurs (Moratalla and Powell 1990). For example, it has been suggested that arcs comprised of 15 to 20 eggs (radii, 1.3 1.7m) resulted from the turning radius of the egg-laying female (Cousin et al. 1990). By comparing published limb dimensions of the European Late Cretaceous titanosaur Hypselosaurus to radii of the arcs, Cousin et al. (1990) proposed a crouching position for the adult sauropod during egg laying. However, none of the eggs assumed to be of Hypselosaurus contain embryonic remains, thus rendering their identification as indeterminate. Such ad hoc assumptions regarding the taxonomy of the egg-laying dinosaur can potentially bias interpretations of data and obscure the true taphonomic picture. Colonial nesting and/or site fidelity have also been hypothesized for sauropod dinosaurs (Sanz et al., 1995; Figueroa and Powell 2000; López-Martínez 2000; Mohabey 2000). One study reported abundant fragmented eggshell and 24 megaloolithid nests arranged in three clusters in a 6,000-m 2 area. Extrapolation of the data, however, extended the number of eggs to 300,000, purportedly laid by sauropods nesting on a seashore (Sanz et al. 1995; López- Martínez 2000). Inferences made from these calculations included territorial behavior, high population density, site fidelity, and site preference (Sanz et al. 1995). These interpretations were challenged on the basis of time-averaging of the deposit, nonsynchronous deposition of egg horizons, pedogenesis, and other sedimentological/taphonomic evidence (Sander et al. 1998). Without detailed taphonomic analysis, such paleobiological inferences appear unwarranted (Sander et al. 1998). Taxonomically unidentified eggs have also been used in studies of sauropod physiology (Case 1978; Erben et al. 1979; Bakker 1986; Paul 1990). For example, egg size data and estimated sauropod hatchling weight were used for determining lifetime reproductive potential as a function of sauropod body mass (Paul 1990) and to estimate the age of sauropods at sexual maturity (Case 1978). In light of the taxonomic uncertainty of the eggs used for these studies, their conclusions are unwarranted. Many megaloolithid eggs found worldwide share a similar structural morphology with NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 299

specimens from Auca Mahuevo and most likely represent eggs laid by sauropod dinosaurs. However, taxonomic identification of eggs based on bones within the same stratigraphic unit carries a high potential for error, as shown with the misidentification of both oviraptorid and troodontid eggs (Norell et al. 1994; Horner and Weishampel 1996). In the case of megaloolithid eggs, this situation is further complicated by the discovery of neonate remains of the hadrosaurid Telmatosaurus transylvanicus in the proximity of megaloolithid egg-clutches from the late Cretaceous of Romania (Grigorescu et al. 1994, 2003), an association that, if confirmed, would highlight the paraphyletic nature of this egg category. Egg studies that depend on taxonomic or ontogenetic comparisons (e.g., embryo to adult size), therefore, should await definitive identification based on embryonic remains within the egg. Thus far, only the in ovo sauropod remains discovered at Auca Mahuevo provide this crucial evidence. ACKNOWLEDGMENTS We thank Michelle Schwengle for rendering the illustrations and Richard Aspinall for assisting with the statistical analyses of egg distribution of Auca Mahuevo s egg-bed 3. We are also grateful to Kristi Curry Rogers and Jeff Wilson for inviting us to contribute to this book and to David Varricchio for his thorough review of the original manuscript. Field and postfield research for this research was supported by the Ann and Gordon Getty Foundation, the Charlotte and Walter Kohler Charitable Trust, the Dirección General de Cultura de Neuquén, the Fundación Antorchas, the Infoquest Foundation, and the Municipalidad de Plaza Huincul. LITERATURE CITED Andreis, R. R. 2001. Paleoecology and environments of the Cretaceous sedimentary basins of Patagonia (southern Argentina). VII International Symposium on Mesozoic Terrestrial Ecosystems. Publ. Espec. Asoc. Paleontol. Argentina 7: 7 14. Ardolino, A. A., and Franchi, M. R. 1996. Geología y Recursos Minerales del Departamento Añelo. Provincia del Neuquén, Rep. Argentina. Publicación Conjunta de la Dirección Provincial de Minería de la Provincia del Neuquén y Dirección Nacional del Servicio Geológico. Anaies (Geol.) 25: 9 106. Bakker, R. T. 1986. The Dinosaur Heresies. William Morrow, New York. 482 pp. Calvo, J. O., Engelland, S., Heredia, S. E., and Salgado, L. 1997. First record of dinosaur eggshells (?Sauropoda Megaloolithidae) from Neuquén, Patagonia, Argentina. GAIA 14: 23 32. Case, T. J. 1978. Speculations on the growth rate and reproduction of some dinosaurs. Paleobiology 4: 320 328. Chiappe, L. M., and Dingus, L. 2001. Walking on Eggs. Scribner, New York. 219 pp. Chiappe, L. M., Coria, R. A., Dingus, L., Jackson, F., Chinsamy, A., and Fox, M. 1998. Sauropod dinosaur embryos from the Late Cretaceous of Patagonia. Nature 396: 258 261. Chiappe, L. M., Dingus, L., Jackson, F., Grellet- Tinner, G., Aspinall, R., Clarke, J., Coria, R. A., Garrido, A., and Loope, D. 2000. Sauropod eggs and embryos from the Upper Cretaceous of Patagonia. 1st Symposium of Dinosaur Eggs and Embryos, Isona, Spain. Pp. 23 29. Chiappe, L. M., Salgado, L., and Coria, R. A. 2001. Embryonic skulls of titanosaur sauropod dinosaurs. Science 293: 2444 2446. Chiappe, L. M., Schmitt, J. G., Jackson, F., Garrido, A., Dingus, L., and Grellet-Tinner, G. 2004. Nest structure for sauropods: Sedimentary criteria for recognition of dinosaur nesting traces. Palaios 19: 89 95. Clark, J. M., Norell, M. A., and Chiappe, L. M. 1999. An oviraptorid skeleton from the Late Cretaceous of Ukhaa Tolgod, Mongolia, preserved in an avian-like brooding position over an oviraptorid nest. Am. Mus. Novitates 3265: 1 36. Coria, R. A., and Arcucci, A. 2005. Nuevos dinosaurios terópodos de Auca Mahuevo, Provincia del Neuquén (Cretácico tardío, Argentina). Ameghiniana (in press). Coria, R. A., and Salgado, L. 1999. Nuevos aportes a la anatomia craneana de los sauropodos titanosauridos. Ameghiniana 36: 98. Coria, R. A., Chiappe, L. M., and Dingus, L. 2002. A new close relative of Carnotaurus sastrei Bonaparte 1985 (Theropoda: Abelisauridae) from the Late Cretaceous of Patagonia. J. Vertebr. Paleontol. 22: 460 465. Cousin, R., 1997. Les gisements d oeufs de dinosauriens des Hautes Corbiéres et des Corbiéres Orientales (Aude): Ponte, nidification, microstructre des coquilles. Bull. Soc. Etudes Sci. Aude 97: 29 46. 300 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES

Cousin, R., Bréton, G., Fournier, R., and Watté, J. 1990. Dinosaur egglaying and nesting in France. In: Carpenter, K., Hirsch, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies. Cambridge University Press, New York. Pp. 56 74. Curry Rogers, K., and Forster., C. 2001. The last of the dinosaur titans: a new sauropod from Madagascar. Nature 412: 530 534.. 2004. The skull of Rapetosaurus krausei (Sauropoda: Titanosauria) from the Late Cretaceous of Madagascar. J. Vertebr. Paleontol. 24: 121 144. Dingus, L., Clarke, J., Scott, G. R., Swisher, C. C., III, Chiappe, L. M., and Coria, R. A. 2000. First magnetostratigraphic/faunal constraints for the age of sauropod embryo-bearing rocks in the Neuquén Group (Late Cretaceous, Neuquén Province, Argentina). Am. Mus. Novitates 3290: 1 11. Dughi, R., and Sirugue, F. 1966. Sur la fossilization des oeufs de dinosaurs. C. R. Sea. Acad. Sci. 262: 2330 2332. Erben, H. 1970. Ultrastrukturen und Mineralisation rezenter und fossiler Eischalen bei Vögeln und Reptilien. Biomineral. Forschungsber. 1: 1 66. Erben, H., Hoefs, J., and Wedepohl, K. H. 1979. Paleobiological and isotopic studies of eggshells from a declining dinosaur species. Paleobiology 5: 380 414. Faccio, G. 1990. Dinosaurian eggs from the Upper Cretaceous of Uruguay. In: Carpenter, K., Hirsch, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies. Cambridge University Press, New York. Pp. 47 55. Figueroa, C., and Powell, J. E. 2000. Structure and permeability properties of thick dinosaur eggshells from the Upper Cretaceous of Patagonia, Argentina. Results of a computer simulation analysis. In: Bravo, A. M., and Reyes, T. (eds.). 1st International Symposium of Dinosaur Eggs and Embryos, Isona, Spain. Pp. 51 60. Garrido, A. C., Chiappe, L. M., Jackson, F., Schmitt, J., and Dingus, L. 2001. First sauropod nest structures. J. Vertebr. Paleontol. 21: 53A. Grellet-Tinner, G., Chiappe, L. M., and Coria, R. A. 2004. Eggs of titanosaurid sauropods from the Upper Cretaceous of Auca Mahuevo (Argentina). Can. J. Earth Sci. 41: 949 960. Grigorescu, D., Weishampel, D., Norman, D., Seclamen, M., Rusu, M., Baltres, A., and Teodorescu, V. 1994. Late Maastrichtian dinosaur eggs from the Hateg Basin (Romania). In: Carpenter, K., Hirsh, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies, Cambridge University Press, Cambridge. Pp. 75 87.. 2003. The puzzle of Tustea incubation site (Upper Maastrichtian, Hateg Basin, Romania): Hadrosaur hatchlings close to Megaloolithidae type of eggshell. 2nd International Symposium on Dinosaur Eggs and Babies, Abstracts, 16. Horner, J. R., and Weishampel, D. B. 1996. A comparative embryological study of two ornithischian dinosaurs A correction. Nature 332: 256 257. Jackson, F. D., Garrido, A. C., and Loope, D. B., 2001. Sauropod egg clutches containing abnormal eggs from the ate Cretaceous of Patagonia: clues to reproductive biology. J. Vertebr. Paleontol. 21: 65A. Jackson, F. D., Garrido, A. C., Schmitt, J., Chiappe, L. M., Dingus, L., and Loope, D. B. 2004. Abnormal, multilayered titanosaurid (Dinosauria: Sauropoda) eggs from in situ clutches at the Auca Mahuevo locality, Neuquén Province, Argentina. J. Vertebr. Paleontol. 24: 913 922. Jacobs, L. L., Winkler, D. A., Downs, W. R., and Gomani, E. 1993. New material of an Early Cretaceous titanosaurid sauropod dinosaur from Malawi. Palaeontology 26: 523 534. Kérourio, P. 1981. Nouvelles observations sur le mode de nidification et de ponte chez les dinosauries du Cretace terminal du Midi de la France. C. R. Sommaire Soc. Geol. France 1: 25 28. Kues, B. S., Lehman, T., and Rigby, J. K. 1980. Teeth of Alamosaurus sanjuanensis, a Late Cretaceous sauropod. J. Paleontol. 54: 864 869. Leanza, H. A. 1999. The Jurassic and Cretaceous terrestrial beds from southern Neuquén Basin, Argentina. Inst. Super. Correl. Geol. Ser. Misc. 4: 1 30. Legarretta, L., and Gulisano, C. 1989. Analisis estratigrafico secuencial de la Cuenca Neuquina (Triasico Superior Terciario Inferior). In: Chebli, G., and Spalletti, L. (eds.). Cuencas Sedimentarias Argentinas, Serie de Correlacion Geologica Nro. 6. Universidad Nacional de Tucumán, Tucumán, Argentina. Pp. 221 243. Loope, D., Schmitt, J., and Jackson, F. 2000. Thunder platters: thin discontinuous limestones in Late Cretaceous continental mudstones of Argentina may be chemically infilled sauropod tracks. Geological Society of America, Abstracts with Programs: A450. López-Martínez, N. 2000. Eggshell sites from the Creatceous-Tertiary transition in south central Pyrenees (Spain). In: Bravo, A. M., and Reyes, T. (eds.). 1st International Symposium of Dinosaur Eggs and Embryos, Isona, Spain. Pp. 95 115. McIntosh, J. 1997. Sauropoda. In: Currie, P. J., and Padian, K. (eds.). Encyclopedia of Dinosaurs, Academic Press, New York. Pp. 654 658. NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES 301

Mikhailov, K. E. 1991. Classification of fossil eggshells of amniotic vertebrates. Acta Paleontol. Polonica 36: 193 238.. 1997. Fossil and recent eggshell in amniotic vertebrates: fine structure, comparative morphology and classification. Special Papers in Paleontology, No. 56. Paleontological Association, London. 80 pp. Mohabey, D. M. 1996. A new oospecies, Megaloolithus matleyi, from the Lameta Formation (Upper Cretaceous) of Chandrapur district, Maharashtra, India, and general remarks on the palaeoenvironment and nesting behavior of dinosaurs. Cretaceous Res. 17: 183 196.. 2000. Indian Upper Cretaceous (Maestrichtian) dinosaur eggs: their parataxonomy and implication in understanding the nesting behavior. In: Bravo, A. M., and Reyes, T. (eds.). 1st International Symposium of Dinosaur Eggs and Embryos, Isona, Spain. Pp. 95 115. Moratalla, J. J., and Powell, J. E. 1990. Dinosaur nesting patterns. In: Carpenter, K., Hirsch, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies. Cambridge University Press, Cambridge. Pp. 37 46. Norell, M. A., Clark, J. M., Dashzeveg, D., Barsbold, R., Chiappe, L. M., Davidson, A., McKenna, M. C., Altangerel, P., and Novacek, M. J. 1994. A theropod dinosaur embryo and the affinities of the Flaming Cliffs dinosaur eggs. Science 266: 779 782. Nowinski, A. 1971. Nemegtosaurus mongoliensis n. gen., n. sp. (Sauropoda) from the Uppermost Cretaceous of Mongolia. Palaeontol. Polonica 25: 57 81. Paul, G. 1990. Dinosaur reproduction in the fast lane: implications for size, success, and extension. In: Carpenter, K., Hirsch, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies. Cambridge University Press, Cambridge. Pp. 244 255. Powell, J. 1992. Hallazgo de huevos asignables a dinosaurios titanosáuridos (Saurischia, Sauropoda) de la Provincia de Río Negro, Argentina. Acta Zool. Lilloana 41: 381 389. Ramos, V. A. 1981. Descripción geológica de la hoja 33c, Los Chihuidos Norte, Provincia del Neuquén. Servicio Nacional de Minería y Geología. Boletín 182. Sahni, A. S. K., Tandon, A., Jolly, S., Bajpai, S., Sood, A., and Srinivasan, S. 1990. Upper Cretaceous dinosaur eggs and nesting sites from the Deccan volcano-sedimentary province of peninsular India. In: Carpenter, K., Hirsch, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies. Cambridge University Press, Cambridge. Pp. 204 226. Salgado, L., and Calvo, J. O. 1997. Evolution of titanosaurid sauropods. II: The cranial evidence. Ameghiniana 34: 33 48. Salgado, L., Coria, R. A., and Chiappe, L. M., 2005. Osteology of the sauropod embryos from the Upper Cretaceous of Argentina. Acta Paleontol Polonica 50: 79 92. Sander, P. M., Peitz, C., Gallemi, J., and Cousin, R. 1998. Dinosaurs nesting on a red beach? Earth Planet. Sci. 327: 67 74. Sanz, J. L., Moratalla, J. J., Díaz-Molina, M., López- Martínez, N., Kälin, O., and Vianey-Liaud, M. 1995. Dinosaur nests at the sea shore. Nature 376: 731 732. Vianey-Liaud, M., Mallan, P., Buscail, O., and C. Montgelard. 1990. Review of French dinosaur eggshells: morphology, structure, mineral, and organic composition. In: Carpenter, K., Hirsch, K. F., and Horner, J. R. (eds.). Dinosaur Eggs and Babies. Cambridge University Press, Cambridge. Pp. 151 183. Vianey-Liaud, M., Hirsch, K. F., Sahni, A., and Sigé, B. 1997. Late Cretaceous Peruvian eggshells and their relationships with Laurasian and Eastern Gondwanian material. Geobios 30: 75 90. Vianey-Liaud, M., Khosla, A., and Garcia, G. 2003. Relationships between European and Indian dinosaur eggs and eggshells of the oofamily Megaloolithidae. J. Vertebr. Paleontol. 23: 575 585. Williams, D. L. G., Seymour, R. S., and Kérourio, P. 1984. Structure of fossil dinosaur eggshell from the Aix Basin, France. Palaeogeogr. Palaeoclimatol. Palaeoecol. 45: 23 37. Wilson, J. A. 2002. Sauropod dinosaur phylogeny: critique and cladistic analysis. Zool. J. Linn. Soc. 136: 217 276. Wilson, J. A., and Sereno, P. C. 1998. Early evolution and higher-level phylogeny of sauropod dinosaurs. J. Vertebr Paleontol 18 (Suppl. 2): 1 68. Zhao, Z. 1979. The advancement of researches on the dinosaurian eggs in China. South China Mesozoic and Cenozoic Red Formation Science, Beijing. Pp. 329 340. 302 NESTING TITANOSAURS FROM AUCA MAHUEVO AND ADJACENT SITES